Competition of Experimentation?
It should be clear to anyone who has studied the topic that the way to drive innovation forward is to have lots of small groups of people each trying different things to succeed. InGuns, Germs, and Steel, for example, we see that certain societies succeed because geography breaks them up into chunks and prevents any one person with bad ideas from getting control of too much, while other societies fail because their whole territory can too easily be captured by an idiot.
It might at first seem more efficient to let the whole territory be captured by a genius, but a moment’s reflection will show that there are few geniuses whose brainpower can match the combined results of many independent experiments. This hasfairly obvious applications to businessand other fields, but for a moment let’s just think about the concept itself.
This idea is often presented as a defense of competition and the capitalist market system that embraces it. Innovation only happens, such people say, when lots of people are competing against each other for the prizes of success. In a communist country, where Big State decides what will be worked on and how, there is no incentive to innovate. Only in a country like ours, where the victor gets the spoils, can new technology be developed.
And yet we also know that competition is a terrible way to get people do well. InNo Contest: The Case Against Competition(now out inan elegant 20th anniversary edition) we see dozens of studies that show that, by all sorts of metrics, people’s performance (and enjoyment) goes down when they are forced to compete. Even worse, it goes down most notably for creative tasks — precisely the kind of thing involved in innovation.
How do we resolve the contradiction? The key is to notice that competition, especially market competition, isn’t the only way to encourage experimentation. And that’s often hard to do, because typically market competition is treated as the only sensible form of competition and competition as the only sensible form of experimentation. But that’s not at all the case.
Instead of providing a prize for winner, we could provide rewards to everyone who tries. And that actually makes sense — not only becauseprizes also decrease productivity and creativity— but also because, when it comes to experimentation, it’s not really your fault if the experiment doesn’t work. In fact, we want to encourage people to try crazy things that might not work, which is exactly why rewards are so counterproductive.
But even if you don’t give an explicit prize, competition is still unhealthy. Contrary to what the apologists for market theology would like you to believe, people do not work better when they’re terrified of the guy next to them finding the solution first. Which is why we should look at this as simply experimentation, not competition.
Experimentation can certainly be carried out cooperatively. Imagine many different scientists in a lab, each trying different ideas during the day, swapping notes and tips over lunch, perhaps joining together to form small groups for certain experiments, or perhaps helping with little pieces of other projects in which they have particular expertise. Each scientist may disagree on which is the right direction to pursue, but that doesn’t make them enemies.
That’s the way that science progresses. And, if you let it, other things too.
You should follow me on twitterhere.
December 7, 2006